The effectiveness of drug applied to a skin surface is dependent on the extent of percutaneous absorption or penetration. For in vitro studies, various diffusion cells have been devised. A cell, comprising a cap, a body and 0-ring, devised by Thomas J. Franz for his study reported in J. Investigative Dermatology, 64, 190 (1975) has been adapted as a commercially available Franz diffusion cell. In this device, the body of the cell constituting the receptor chamber is that portion containing a physiological solution into which a test drug diffuses or penetrates through a test membrane which oftentimes is skin. Samples are withdrawn from the receptor chamber through a needle inserted into a sampling port constituting a side arm open to the atmosphere. The available cells have certain limitations. Thus, sample size is limited to the volume of fluid in the side arm. Moreover, when withdrawal of a large sample is attempted, there is a tendency for air to enter into the receptor chamber. Air thus introduced collects under the membrane being tested, reducing the effective surface area for drug diffusion, and thereby giving rise to erroneous measurements. It is desirable to provide for a diffusion cell which overcomes these limitations.